Vector Quantile Regression on Manifolds

TL;DR

This paper introduces a novel framework for vector quantile regression on manifolds using optimal transport theory, enabling distribution-free estimation of conditional quantiles for data on non-Euclidean spaces like spheres and tori.

Contribution

It extends quantile regression to multivariate distributions on manifolds by defining conditional vector quantile functions using c-concave functions and optimal transport.

Findings

Effective quantile estimation on manifolds demonstrated with synthetic data.

Provides tools for regression, confidence sets, and likelihoods in non-Euclidean spaces.

Insights into the interpretation of non-Euclidean quantiles.

Abstract

Quantile regression (QR) is a statistical tool for distribution-free estimation of conditional quantiles of a target variable given explanatory features. QR is limited by the assumption that the target distribution is univariate and defined on an Euclidean domain. Although the notion of quantiles was recently extended to multi-variate distributions, QR for multi-variate distributions on manifolds remains underexplored, even though many important applications inherently involve data distributed on, e.g., spheres (climate and geological phenomena), and tori (dihedral angles in proteins). By leveraging optimal transport theory and c-concave functions, we meaningfully define conditional vector quantile functions of high-dimensional variables on manifolds (M-CVQFs). Our approach allows for quantile estimation, regression, and computation of conditional confidence sets and likelihoods. We demonstrate the approach's efficacy and provide insights regarding the meaning of non-Euclidean quantiles through synthetic and real data experiments.